**** BEGIN LOGGING AT Fri Nov 29 02:59:58 2019 Nov 29 07:28:16 Morning! Nov 29 08:08:52 Morning! Nov 29 08:29:14 Tofe: Disabling the neva-media doesn't really seem to help either, same problem :S Nov 29 08:57:02 mmh ok Nov 29 08:57:18 I'm not very surprised either, but well Nov 29 12:31:01 Tofe: Yeah we probably need LSM for things to work properly Nov 29 12:43:43 Yes... but the effort needed to bring it to qtwayland 5.12 might be the same as implementing wl_webos_shell on our luna-next... Nov 29 12:44:02 plus, that effort has already been made once Nov 29 12:44:48 but maybe implementing wl_webos_shell in a minimalistic way isn't that big challenge, I could try it Nov 29 12:59:24 Tofe: YOu mean this work that was done for 5.8 migration at our end? https://github.com/webOS-ports/luna-next/commit/30534a4c645e074c2154e72cca7743cabf3d2124 Nov 29 12:59:30 That seems like quite a bit yes Nov 29 13:13:23 yes, this kind of work Nov 29 13:15:37 plus, I'm sure LG already has done it somewhere :p Nov 29 13:16:41 but implementing wl_webos_shell on luna-next, in a minimal way, might be a good idea to smoothen the transition Nov 29 13:17:19 it wouldn't interfere with the other shell either, because the Qt windows won't be using that unknown shell Nov 29 13:17:41 I'll give it a shot this week-end Nov 29 13:17:55 Yeah it shouldn't be too hard probably Nov 29 13:18:10 It's all pretty separated in files etc Nov 29 13:18:27 So should be easy to lift & drop and modify some of our existing files to deal with it Nov 29 13:19:28 yup Nov 29 13:47:17 Herrie: btw, https://www.webosose.org/docs/guides/core-topics/graphics-input/graphics-input-overview/ is a quiite nice overview Nov 29 13:59:41 Yes pretty nice, just an old screenshot on top ;) Nov 29 14:23:17 JaMa: http://jenkins.nas-admin.org/job/LuneOS/view/testing/job/luneos-testing_pinephone/52/console exited with code 1, but all succeeded ? what's the catch ? Nov 29 14:25:35 I can only see some setscene errors, which seem non fatal Nov 29 14:27:21 Tofe: Infra has been acting up a bit recently Nov 29 14:27:30 PRSERV is down, webos-ports.org is down Nov 29 14:28:43 ah, ok Nov 29 14:30:37 Sent an email to Tom, Nick & Derek. Maybe they were doing maintenance or didn't notice ;) Nov 29 14:49:37 Tofe: sstate from milla sometimes fails to fetch Nov 29 14:50:32 ok Nov 29 14:50:33 Tofe: first it's there when bitbake checks for its existence, but then it drops the connection when fetching the archive itself, bitbake shows error, but then runs the task itself (instead of reusing sstate) so in the end everything is OK, the error code is just from those ERROR: messages about disappeared sstate Nov 29 14:50:57 it's failing like this quite often on jenkins (when the sstate isn't all local on bonaire) Nov 29 14:51:34 isn't failing on sstate miss a bit harsh ? if runqueue was fine, I mean Nov 29 14:52:42 yes, that's why I still carry these 2 patches in my forks https://github.com/shr-distribution/bitbake/commit/59c8eafcf5e31ea976818257189d91ce79b47d35 http://git.openembedded.org/openembedded-core-contrib/commit/?h=jansa/master&id=5336dd23a9fda79c9ef32ceed340c02703766982 Nov 29 14:53:03 ah, nice :) Nov 29 14:53:07 but it was already rejected 2 times, so it probably won't get accepted Nov 29 14:54:29 RP wants to show the ERROR, because these issue say something about the infra, but I don't want builds reporting failure like this (and I don't want to grep the build output to find out if it was only these setscene errors or something else was wrong as well - but maybe I should just add this to jenkins-job.sh eventually) Nov 29 14:55:15 so currently I'm just re-triggering the failed jobs, so that you don't think something is broken - except sometimes I'm a bit late and you do notice before it's rebuilt :) Nov 29 14:57:18 ahah, damn, you do that manually Nov 29 14:58:54 and I guess there's no good way of patching bitbake on jenkins without interfering with updates Nov 29 15:08:59 if we have to patch it, then I would use our fork in webos-ports-setup (instead of patching it automagically during the build and dealing with the patch failing to apply sometimes) Nov 29 15:13:45 JaMa: I don't see why they would have issues merging this when it's adjustable by a flag.... Nov 29 15:24:24 Herrie: you can find the discussion on the ML, but in short RP doesn't want more flags and code paths Nov 29 15:26:56 Let's just remove the flag and change the default behavior then ;) Nov 29 15:31:54 he wants the build to fail when this happens Nov 29 15:32:43 :( ok Nov 29 15:36:49 In an ideal world with extremely reliable hardware: Yes, but that's not the case everywhere ;) Nov 29 16:27:18 something like this should work https://github.com/webOS-ports/jenkins-jobs/pull/12 but it's really ugly Nov 29 16:29:00 https://marc.info/?l=openembedded-core&m=150408679301031&w=2 Nov 29 16:33:47 and https://marc.info/?l=openembedded-core&m=150407969131099&w=2 Nov 29 16:49:32 and my rant in the reply to this old thread now: http://lists.openembedded.org/pipermail/openembedded-core/2019-November/289901.html Nov 29 19:49:31 Herrie: completely totally untested (just builds), but if you're feeling adventurous you can try https://github.com/webOS-ports/luna-next/commit/b11a2ac9201619612e597989483773e2dbe72c32 Nov 29 19:54:19 JaMa: yep, ugly, but... does the job... overall we're saving time Nov 29 19:55:56 Tofe: ooh nice Nov 29 19:56:12 Will give that a go probably early morning Nov 29 19:58:01 Corporate party tonight & a bike to pick up & fix ;) Nov 29 19:58:30 Seems brakes jammed again on the disc :'( Time for new caliper I guess Nov 29 19:59:42 It's been a bit finicky since start, so better to replace I guess **** ENDING LOGGING AT Sat Nov 30 02:59:57 2019